On Missing Labels, Long-tails and Propensities in Extreme Multi-label Classification.
Erik SchultheisMarek WydmuchRohit BabbarKrzysztof DembczynskiPublished in: KDD (2022)
Keyphrases
- multi label classification
- multi label
- label space
- image annotation
- semi supervised learning
- text categorization
- binary classification
- multiple labels
- binary classifiers
- classification algorithm
- semi supervised
- graph cuts
- single label classification
- meta level
- multi label learning
- learning tasks
- text classification
- data sets
- multi class
- class labels
- high level
- support vector
- low level
- data points
- image processing
- classification accuracy
- cost sensitive
- knn
- support vector machine
- image classification
- labeled data
- k nearest neighbor